Cleanup set info. (#10139)

- Use the array interface internally. - Deprecate `XGDMatrixSetDenseInfo`. - Deprecate `XGDMatrixSetUIntInfo`. - Move the handling of `DataType` into the deprecated C function. --------- Co-authored-by: Philip Hyunsu Cho <chohyu01@cs.washington.edu>
dmlc · Mar 26, 2024 · 230010d · 230010d
1 parent 6a7c6a8
commit 230010d
Show file tree

Hide file tree

Showing 37 changed files with 245 additions and 267 deletions.
diff --git a/.github/workflows/r_tests.yml b/.github/workflows/r_tests.yml
@@ -110,7 +110,7 @@ jobs:
     name: Test R package on Debian
     runs-on: ubuntu-latest
     container:
-      image: rhub/debian-gcc-devel
+      image: rhub/debian-gcc-release
 
     steps:
     - name: Install system dependencies
@@ -130,12 +130,12 @@ jobs:
     - name: Install dependencies
       shell: bash -l {0}
       run: |
-        /tmp/R-devel/bin/Rscript -e "source('./R-package/tests/helper_scripts/install_deps.R')"
+        Rscript -e "source('./R-package/tests/helper_scripts/install_deps.R')"
 
     - name: Test R
       shell: bash -l {0}
       run: |
-        python3 tests/ci_build/test_r_package.py --r=/tmp/R-devel/bin/R --build-tool=autotools --task=check
+        python3 tests/ci_build/test_r_package.py --r=/usr/bin/R --build-tool=autotools --task=check
 
     - uses: dorny/paths-filter@v2
       id: changes
@@ -147,4 +147,4 @@ jobs:
     - name: Run document check
       if: steps.changes.outputs.r_package == 'true'
       run: |
-        python3 tests/ci_build/test_r_package.py --r=/tmp/R-devel/bin/R --task=doc
+        python3 tests/ci_build/test_r_package.py --r=/usr/bin/R --task=doc
diff --git a/include/xgboost/c_api.h b/include/xgboost/c_api.h
@@ -1,5 +1,5 @@
 /**
- * Copyright 2015~2023 by XGBoost Contributors
+ * Copyright 2015-2024, XGBoost Contributors
  * \file c_api.h
  * \author Tianqi Chen
  * \brief C API of XGBoost, used for interfacing to other languages.
@@ -639,21 +639,14 @@ XGB_DLL int XGDMatrixSetInfoFromInterface(DMatrixHandle handle,
  * \param len length of array
  * \return 0 when success, -1 when failure happens
  */
-XGB_DLL int XGDMatrixSetFloatInfo(DMatrixHandle handle,
-                                  const char *field,
-                                  const float *array,
+XGB_DLL int XGDMatrixSetFloatInfo(DMatrixHandle handle, const char *field, const float *array,
                                   bst_ulong len);
-/*!
- * \brief set uint32 vector to a content in info
- * \param handle a instance of data matrix
- * \param field field name
- * \param array pointer to unsigned int vector
- * \param len length of array
- * \return 0 when success, -1 when failure happens
+/**
+ * @deprecated since 2.1.0
+ *
+ * Use @ref XGDMatrixSetInfoFromInterface instead.
  */
-XGB_DLL int XGDMatrixSetUIntInfo(DMatrixHandle handle,
-                                 const char *field,
-                                 const unsigned *array,
+XGB_DLL int XGDMatrixSetUIntInfo(DMatrixHandle handle, const char *field, const unsigned *array,
                                  bst_ulong len);
 
 /*!
@@ -725,42 +718,13 @@ XGB_DLL int XGDMatrixGetStrFeatureInfo(DMatrixHandle handle, const char *field,
                                        bst_ulong *size,
                                        const char ***out_features);
 
-/*!
- * \brief Set meta info from dense matrix.  Valid field names are:
- *
- *  - label
- *  - weight
- *  - base_margin
- *  - group
- *  - label_lower_bound
- *  - label_upper_bound
- *  - feature_weights
+/**
+ * @deprecated since 2.1.0
  *
- * \param handle An instance of data matrix
- * \param field  Field name
- * \param data   Pointer to consecutive memory storing data.
- * \param size   Size of the data, this is relative to size of type.  (Meaning NOT number
- *               of bytes.)
- * \param type   Indicator of data type.  This is defined in xgboost::DataType enum class.
- *    - float    = 1
- *    - double   = 2
- *    - uint32_t = 3
- *    - uint64_t = 4
- * \return 0 when success, -1 when failure happens
- */
-XGB_DLL int XGDMatrixSetDenseInfo(DMatrixHandle handle, const char *field,
-                                  void const *data, bst_ulong size, int type);
-
-/*!
- * \brief (deprecated) Use XGDMatrixSetUIntInfo instead. Set group of the training matrix
- * \param handle a instance of data matrix
- * \param group pointer to group size
- * \param len length of array
- * \return 0 when success, -1 when failure happens
+ * Use @ref XGDMatrixSetInfoFromInterface instead.
  */
-XGB_DLL int XGDMatrixSetGroup(DMatrixHandle handle,
-                              const unsigned *group,
-                              bst_ulong len);
+XGB_DLL int XGDMatrixSetDenseInfo(DMatrixHandle handle, const char *field, void const *data,
+                                  bst_ulong size, int type);
 
 /*!
  * \brief get float info vector from matrix.

diff --git a/include/xgboost/data.h b/include/xgboost/data.h
@@ -19,7 +19,6 @@
 #include <algorithm>
 #include <limits>
 #include <memory>
-#include <numeric>
 #include <string>
 #include <utility>
 #include <vector>
@@ -137,14 +136,6 @@ class MetaInfo {
    * \param fo The output stream.
    */
   void SaveBinary(dmlc::Stream* fo) const;
-  /*!
-   * \brief Set information in the meta info.
-   * \param key The key of the information.
-   * \param dptr The data pointer of the source array.
-   * \param dtype The type of the source data.
-   * \param num Number of elements in the source array.
-   */
-  void SetInfo(Context const& ctx, const char* key, const void* dptr, DataType dtype, size_t num);
   /*!
    * \brief Set information in the meta info with array interface.
    * \param key The key of the information.
@@ -517,10 +508,6 @@ class DMatrix {
   DMatrix()  = default;
   /*! \brief meta information of the dataset */
   virtual MetaInfo& Info() = 0;
-  virtual void SetInfo(const char* key, const void* dptr, DataType dtype, size_t num) {
-    auto const& ctx = *this->Ctx();
-    this->Info().SetInfo(ctx, key, dptr, dtype, num);
-  }
   virtual void SetInfo(const char* key, std::string const& interface_str) {
     auto const& ctx = *this->Ctx();
     this->Info().SetInfo(ctx, key, StringView{interface_str});

diff --git a/include/xgboost/linalg.h b/include/xgboost/linalg.h
@@ -190,13 +190,14 @@ constexpr auto ArrToTuple(T (&arr)[N]) {
 // uint division optimization inspired by the CIndexer in cupy.  Division operation is
 // slow on both CPU and GPU, especially 64 bit integer.  So here we first try to avoid 64
 // bit when the index is smaller, then try to avoid division when it's exp of 2.
-template <typename I, int32_t D>
+template <typename I, std::int32_t D>
 LINALG_HD auto UnravelImpl(I idx, common::Span<size_t const, D> shape) {
-  size_t index[D]{0};
+  std::size_t index[D]{0};
   static_assert(std::is_signed<decltype(D)>::value,
                 "Don't change the type without changing the for loop.");
+  auto const sptr = shape.data();
   for (int32_t dim = D; --dim > 0;) {
-    auto s = static_cast<std::remove_const_t<std::remove_reference_t<I>>>(shape[dim]);
+    auto s = static_cast<std::remove_const_t<std::remove_reference_t<I>>>(sptr[dim]);
     if (s & (s - 1)) {
       auto t = idx / s;
       index[dim] = idx - t * s;
@@ -745,6 +746,14 @@ auto ArrayInterfaceStr(TensorView<T, D> const &t) {
   return str;
 }
 
+template <typename T>
+auto Make1dInterface(T const *vec, std::size_t len) {
+  Context ctx;
+  auto t = linalg::MakeTensorView(&ctx, common::Span{vec, len}, len);
+  auto str = linalg::ArrayInterfaceStr(t);
+  return str;
+}
+
 /**
  * \brief A tensor storage. To use it for other functionality like slicing one needs to
  *        obtain a view first.  This way we can use it on both host and device.

diff --git a/include/xgboost/span.h b/include/xgboost/span.h
@@ -30,9 +30,8 @@
 #define XGBOOST_SPAN_H_
 
 #include <xgboost/base.h>
-#include <xgboost/logging.h>
 
-#include <cinttypes>  // size_t
+#include <cstddef>  // size_t
 #include <cstdio>
 #include <iterator>
 #include <limits>  // numeric_limits
@@ -73,8 +72,7 @@
 
 #endif  // defined(_MSC_VER) && _MSC_VER < 1910
 
-namespace xgboost {
-namespace common {
+namespace xgboost::common {
 
 #if defined(__CUDA_ARCH__)
 // Usual logging facility is not available inside device code.
@@ -707,8 +705,8 @@ class IterSpan {
     return it_ + size();
   }
 };
-}  // namespace common
-}  // namespace xgboost
+}  // namespace xgboost::common
+
 
 #if defined(_MSC_VER) &&_MSC_VER < 1910
 #undef constexpr

diff --git a/jvm-packages/xgboost4j/src/native/xgboost4j.cpp b/jvm-packages/xgboost4j/src/native/xgboost4j.cpp
@@ -408,7 +408,8 @@ JNIEXPORT jint JNICALL Java_ml_dmlc_xgboost4j_java_XGBoostJNI_XGDMatrixSetFloatI
 
   jfloat* array = jenv->GetFloatArrayElements(jarray, NULL);
   bst_ulong len = (bst_ulong)jenv->GetArrayLength(jarray);
-  int ret = XGDMatrixSetFloatInfo(handle, field, (float const *)array, len);
+  auto str = xgboost::linalg::Make1dInterface(array, len);
+  int ret = XGDMatrixSetInfoFromInterface(handle, field, str.c_str());
   JVM_CHECK_CALL(ret);
   //release
   if (field) jenv->ReleaseStringUTFChars(jfield, field);
@@ -427,7 +428,8 @@ JNIEXPORT jint JNICALL Java_ml_dmlc_xgboost4j_java_XGBoostJNI_XGDMatrixSetUIntIn
   const char*  field = jenv->GetStringUTFChars(jfield, 0);
   jint* array = jenv->GetIntArrayElements(jarray, NULL);
   bst_ulong len = (bst_ulong)jenv->GetArrayLength(jarray);
-  int ret = XGDMatrixSetUIntInfo(handle, (char const *)field, (unsigned int const *)array, len);
+  auto str = xgboost::linalg::Make1dInterface(array, len);
+  int ret = XGDMatrixSetInfoFromInterface(handle, field, str.c_str());
   JVM_CHECK_CALL(ret);
   //release
   if (field) jenv->ReleaseStringUTFChars(jfield, (const char *)field);
@@ -730,8 +732,8 @@ JNIEXPORT jint JNICALL Java_ml_dmlc_xgboost4j_java_XGBoostJNI_XGBoosterPredictFr
   if (jmargin) {
     margin = jenv->GetFloatArrayElements(jmargin, nullptr);
     JVM_CHECK_CALL(XGProxyDMatrixCreate(&proxy));
-    JVM_CHECK_CALL(
-        XGDMatrixSetFloatInfo(proxy, "base_margin", margin, jenv->GetArrayLength(jmargin)));
+    auto str = xgboost::linalg::Make1dInterface(margin, jenv->GetArrayLength(jmargin));
+    JVM_CHECK_CALL(XGDMatrixSetInfoFromInterface(proxy, "base_margin", str.c_str()));
   }
 
   bst_ulong const *out_shape;

diff --git a/src/c_api/c_api.cc b/src/c_api/c_api.cc
@@ -1,5 +1,5 @@
 /**
- * Copyright 2014-2024 by XGBoost Contributors
+ * Copyright 2014-2024, XGBoost Contributors
  */
 #include "xgboost/c_api.h"
 
@@ -614,8 +614,8 @@ XGB_DLL int XGDMatrixSetFloatInfo(DMatrixHandle handle, const char *field, const
   API_BEGIN();
   CHECK_HANDLE();
   xgboost_CHECK_C_ARG_PTR(field);
-  auto const& p_fmat = *static_cast<std::shared_ptr<DMatrix> *>(handle);
-  p_fmat->SetInfo(field, info, xgboost::DataType::kFloat32, len);
+  auto const &p_fmat = *static_cast<std::shared_ptr<DMatrix> *>(handle);
+  p_fmat->SetInfo(field, linalg::Make1dInterface(info, len));
   API_END();
 }
 
@@ -634,8 +634,9 @@ XGB_DLL int XGDMatrixSetUIntInfo(DMatrixHandle handle, const char *field, const
   API_BEGIN();
   CHECK_HANDLE();
   xgboost_CHECK_C_ARG_PTR(field);
+  LOG(WARNING) << error::DeprecatedFunc(__func__, "2.1.0", "XGDMatrixSetInfoFromInterface");
   auto const &p_fmat = *static_cast<std::shared_ptr<DMatrix> *>(handle);
-  p_fmat->SetInfo(field, info, xgboost::DataType::kUInt32, len);
+  p_fmat->SetInfo(field, linalg::Make1dInterface(info, len));
   API_END();
 }
 
@@ -679,19 +680,52 @@ XGB_DLL int XGDMatrixSetDenseInfo(DMatrixHandle handle, const char *field, void
                                   xgboost::bst_ulong size, int type) {
   API_BEGIN();
   CHECK_HANDLE();
+  LOG(WARNING) << error::DeprecatedFunc(__func__, "2.1.0", "XGDMatrixSetInfoFromInterface");
   auto const &p_fmat = *static_cast<std::shared_ptr<DMatrix> *>(handle);
   CHECK(type >= 1 && type <= 4);
   xgboost_CHECK_C_ARG_PTR(field);
-  p_fmat->SetInfo(field, data, static_cast<DataType>(type), size);
-  API_END();
-}
 
-XGB_DLL int XGDMatrixSetGroup(DMatrixHandle handle, const unsigned *group, xgboost::bst_ulong len) {
-  API_BEGIN();
-  CHECK_HANDLE();
-  LOG(WARNING) << "XGDMatrixSetGroup is deprecated, use `XGDMatrixSetUIntInfo` instead.";
-  auto const &p_fmat = *static_cast<std::shared_ptr<DMatrix> *>(handle);
-  p_fmat->SetInfo("group", group, xgboost::DataType::kUInt32, len);
+  Context ctx;
+  auto dtype = static_cast<DataType>(type);
+  std::string str;
+  auto proc = [&](auto cast_d_ptr) {
+    using T = std::remove_pointer_t<decltype(cast_d_ptr)>;
+    auto t = linalg::TensorView<T, 1>(
+        common::Span<T>{cast_d_ptr, static_cast<typename common::Span<T>::index_type>(size)},
+        {size}, DeviceOrd::CPU());
+    CHECK(t.CContiguous());
+    Json interface{linalg::ArrayInterface(t)};
+    CHECK(ArrayInterface<1>{interface}.is_contiguous);
+    str = Json::Dump(interface);
+    return str;
+  };
+
+  // Legacy code using XGBoost dtype, which is a small subset of array interface types.
+  switch (dtype) {
+    case xgboost::DataType::kFloat32: {
+      auto cast_ptr = reinterpret_cast<const float *>(data);
+      p_fmat->Info().SetInfo(ctx, field, proc(cast_ptr));
+      break;
+    }
+    case xgboost::DataType::kDouble: {
+      auto cast_ptr = reinterpret_cast<const double *>(data);
+      p_fmat->Info().SetInfo(ctx, field, proc(cast_ptr));
+      break;
+    }
+    case xgboost::DataType::kUInt32: {
+      auto cast_ptr = reinterpret_cast<const uint32_t *>(data);
+      p_fmat->Info().SetInfo(ctx, field, proc(cast_ptr));
+      break;
+    }
+    case xgboost::DataType::kUInt64: {
+      auto cast_ptr = reinterpret_cast<const uint64_t *>(data);
+      p_fmat->Info().SetInfo(ctx, field, proc(cast_ptr));
+      break;
+    }
+    default:
+      LOG(FATAL) << "Unknown data type" << static_cast<uint8_t>(dtype);
+  }
+
   API_END();
 }
 
@@ -987,7 +1021,7 @@ XGB_DLL int XGBoosterBoostOneIter(BoosterHandle handle, DMatrixHandle dtrain, bs
                                   bst_float *hess, xgboost::bst_ulong len) {
   API_BEGIN();
   CHECK_HANDLE();
-  error::DeprecatedFunc(__func__, "2.1.0", "XGBoosterTrainOneIter");
+  LOG(WARNING) << error::DeprecatedFunc(__func__, "2.1.0", "XGBoosterTrainOneIter");
   auto *learner = static_cast<Learner *>(handle);
   auto ctx = learner->Ctx()->MakeCPU();
 

diff --git a/src/c_api/c_api_utils.h b/src/c_api/c_api_utils.h
@@ -1,17 +1,18 @@
 /**
- * Copyright 2021-2023, XGBoost Contributors
+ * Copyright 2021-2024, XGBoost Contributors
  */
 #ifndef XGBOOST_C_API_C_API_UTILS_H_
 #define XGBOOST_C_API_C_API_UTILS_H_
 
-#include <algorithm>
-#include <cstddef>
-#include <functional>
-#include <memory>   // for shared_ptr
-#include <string>   // for string
-#include <tuple>    // for make_tuple
-#include <utility>  // for move
-#include <vector>
+#include <algorithm>   // for min
+#include <cstddef>     // for size_t
+#include <functional>  // for multiplies
+#include <memory>      // for shared_ptr
+#include <numeric>     // for accumulate
+#include <string>      // for string
+#include <tuple>       // for make_tuple
+#include <utility>     // for move
+#include <vector>      // for vector
 
 #include "../common/json_utils.h"  // for TypeCheck
 #include "xgboost/c_api.h"

diff --git a/src/collective/nccl_device_communicator.cu b/src/collective/nccl_device_communicator.cu
@@ -2,6 +2,8 @@
  * Copyright 2023 XGBoost contributors
  */
 #if defined(XGBOOST_USE_NCCL)
+#include <numeric>  // for accumulate
+
 #include "comm.cuh"
 #include "nccl_device_communicator.cuh"
 

diff --git a/src/common/error_msg.cc b/src/common/error_msg.cc
@@ -11,7 +11,7 @@
 #include "xgboost/logging.h"
 
 namespace xgboost::error {
-std::string DeprecatedFunc(StringView old, StringView since, StringView replacement) {
+[[nodiscard]] std::string DeprecatedFunc(StringView old, StringView since, StringView replacement) {
   std::stringstream ss;
   ss << "`" << old << "` is deprecated since" << since << ", use `" << replacement << "` instead.";
   return ss.str();

diff --git a/src/common/error_msg.h b/src/common/error_msg.h
@@ -89,7 +89,7 @@ void WarnDeprecatedGPUId();
 
 void WarnEmptyDataset();
 
-std::string DeprecatedFunc(StringView old, StringView since, StringView replacement);
+[[nodiscard]] std::string DeprecatedFunc(StringView old, StringView since, StringView replacement);
 
 constexpr StringView InvalidCUDAOrdinal() {
   return "Invalid device. `device` is required to be CUDA and there must be at least one GPU "